Magnetic resonance imaging (MRI) is a medical imaging modality that can create pictures of the inside of a human body without using x-rays or other ionizing radiation. MRI data may be acquired as three-dimensional (3-D) data-sets and displayed as two-dimensional (2-D) images corresponding to individual slices through the 3-D data-set. Alternatively, a 3-D MRI data-set may be displayed as a 2-D image using a format that seeks to represent objects of interest in the data-set from the perspective of a hypothetical external observer. Examples of methods for generating a 2-D image from a 3-D MRI data set include “volume-rendering” and “surface-rendering” techniques.
In volume-rendering techniques, each “voxel” (or volume element) of a 3-D data-set is assigned a color and opacity based on its image intensity. Rays are then cast from a virtual camera at the location of a theoretical observer into the 3-D image data, with each ray generating a single pixel in the resultant 2-D image. The color produced by the ray at the resultant pixel is determined from the color and opacity values of the voxels encountered along its trace.
In surface-rendering techniques, the 3-D image data is first “segmented”, i.e., each voxel in the data-set is assigned a label corresponding to a predefined tissue type. Then, a polygonal mesh is wrapped around the segmented tissue structures to form surface models for the structures. Typically, each surface is displayed using a single color and opacity that are assigned on the basis of the corresponding tissue type. Alternatively, a color for a surface may be determined based on measured surface properties such as curvature and thickness.
For many clinical applications, depicting a 3-D MRI data-set using such known techniques is sufficient. However, advanced applications such as MRI-guided surgery or MRI-assisted endoscopy may benefit from more sophisticated image display techniques. It may be useful to provide improved renderings of surfaces (e.g., organ surfaces, skin surfaces) in such applications to aid in, for example, surgical planning. Accordingly, it would be desirable to provide a method and apparatus for providing more accurate surface renderings for MRI images. In particular, it would be desirable to provide a method and apparatus for combining photographic image data with 3-D MRI data.